January 27, 2013 - 7:05 pm. Posted by Plamen Ganchosov
With 2.5 quintillion bytes of data created every day, it’s not surprising companies are worried about the advanced computer infrastructure required to handle their data-intensive workloads. Such infrastructure is not only expensive, but also difficult to maintain. Luckily, the answer to this concern may lie in today’s most innovative technology environment: cloud computing.
By taking advantage of powerful cloud computing platforms, companies can access the computing resources they need at a competitive cost and without the need to constantly procure and scale cumbersome in-house IT infrastructures. To handle the unpredictable nature of research science and the high volumes of data produced on a daily basis, it’s essential to have a cloud platform that places no restrictions on server sizes, software or networking, and offers a fully scalable and customizable infrastructure.
First, on January 29, 2013 from 3:35-4:00 p.m. GMT, Bernino will be joined by Ian Bird, CERN’s Worldwide LHC Computing Grid Project Leader, on the session: CERN: Big Science Data and the Cloud. The presentation will discuss how CERN has built a worldwide distributed computing environment to support analysis of its 70 PB of LHC data. As part of the “Helix Nebula – The Science Cloud” initiative, CERN has been moving parts of this environment into the cloud. The talk will describe some of the challenges and successes, and will discuss the outlook for scientific big data processing in clouds.
Then, on January 30, 2013 from 10:55-11:20 a.m. GMT, Bernino will present the session, Big Data in a public cloud: Maximize RAM without paying for overhead CPU and other price/performance tricks.This presentation will cover practical and pragmatic considerations for deploying big data infrastructure in a public cloud. It will also detail how to optimize deployments, efficient purchasing strategies and hacks that make big data workloads fly in the cloud. The presentation is founded on real world examples from big data customers from vertical industries that use CloudSigma to process their heavy workloads.